Oh, Ghod; I hate this whole "the desktop has no edges" paradigm... (Of course, I hate the word 'paradigm', too, but what the hell...:-)
Am I the only one who thinks this is a _bad_ thing? People want limits, they want walls; this has been proven so many times I can't count them.
Certainly, those walls should not _preclude_ you from finding and making use of the things you need, but really, now; do you _want_ to be unable to discern where that love note to your amor at work... or your HIV test results are; whether that file is on your zip disk, your hard drive, your office fileserver... or just 'out there on the web somewhere'?
Really; c'mon now. Reductio ad absurdum, maybe, but look at it the other way: your web browser, properly configured, is actually your generic file viewer, whether or not those files are _local_, right?
I often take Alan Cooper to task (although not nearly often enough in public:-) for his complete ignorance of the installed base in proposing things like a complete overhaul of the Common User Interface inspired guidelines (such as they are) for Windows apps, and for advocating completely hiding the hierarchical filesystem -- and indeed, the filesystem in general.
These might be good ideas, but I simply don't believe that it's practical to implement them in an environment where "everybody else's" Windows program works some other way.
(This is completely apart from my opinion that these things comprise fundamental computer literacy and that, contrary to Cooper's assertions in Inmates, computer literacy is not a polite way of saying 'stupid willingness to tolerate insufferably poor design -- I don't think HFS' are poor design.)
But, my question would be (yes, there's actually a question buried in here) this: you, Jakob, seem to go the other direction, displaying an almost slavish devotion to the installed base, even if a) it's a poor design, objectively, and b) there's an obviously better one.
Color choices for links would be my favorite example. My weblog, for example, uses a dark teal text color (dark enough for good contrast on white), and navy blue for links -- which drops back to the text color, while still staying underlined, for visited links.
Obviously, this wouldn't work on navigational elements rather than in-line links, but I'm honestly not at all certain that visited-ness is pertinent there, and it might actually be confusing. (I mean, here, the sort of things many people put on 'tabs'.)
I'm an engineering type, so I suppose I'd rather have it 'right', even if it means I have to work harder training people the first time, because if I teach them to fish (discerning rules for link changes from looking at a site) rtaher than feeding them (blue means link, red means visited), then my long-term support load decreases.
Do you think that people, in general, are really so untrainable that this is an unreasonable approach? Or that making them think, even just a little bit, is really that much to ask?
The Clydesdale ad goes in the "Aw..." category, in which category it won.
Without chasing everyone over to my weblog for the grind, the favorite in the room I was watching in was the Tabasco spot (the 'log has a link to the QT4 file), followed by the Clydesdale spot, the FedEx/Wizard of Oz spot, and the VISA 'swimmers' spot, in differing orders of preference.
We did like the Christopher Reeve one, too. And the "slidewalk/credit card fraud" company's stuff sounded pretty cool, too... who were they again?:-)
I, personally, was quite impressed by the 7 books of the "Incarnations of Immortality" series -- primarily because I am _not_ a fantasy reader, by and large. I don't much mind the "Wizard" series from Rick Cook because of all the nifty computing jokes he slips in -- being a long time writer for Byte as well (Toth Set Ra, indeed!).
But the Incarnations, especially books 6 and 7 when he finally got them written, were just hard enough to make me happy.
> You know there's actually a law in Florida from the turn of the Century (still in effect) that goes something like:
"When a person driving a motor vehicle reaches a horse and the horse acts frightened, the driver of the vehicle must get out of the car and take it apart, laying its pieces in front of the horse so the horse will calm down"
Um, like hell there is.
I have a set of Florida lawbooks on my office shelf. I've read every word of chapters 316-321, which cover motor vehicles, driving, licensure and titles. Nothing like this exists anywhere in that section.
You have a citation?
(Yeah, I know, he was just trying to make a point... with cobbled-up facts. Furrfu!)
The original writer notes: > But this is not an OS issue, it never was.
I think he's wrong, and our much more than usually eloquent Anonymous Coward appears to agree: > We can't all work together, because some of "us" depend for "our" existence on keeping code proprietary, and some of us depend on keeping it free. Microsoft can't compromise with free-software, because if it does, it effectively dies as Microsoft. Sure, it could exist as a distro company for FreeWindows2000, but it would no longer be Microsoft in anything but name.
And this comes very close to making the point, for me. It was, roughly, would an NT user/programmer have done this for Microsoft? Or, even moreso, for a Linux company?
This is the same argument, really, as "oh, he's a fashion designer, he must be gay."
No, being gay doesn't make you a better fashion designer. However, _having been gay_ may well predispose you to a greater disdain for society's opinion of whether men should engage in such pursuits. Correlation doesn't imply causation, no. However, that is _not_ sufficient reason to deny such correlations.
The mindset that makes someone likely to hack on Linux, or even use it (still, Corel isn't on 40 million dekstops _yet) is one that's more likely to yield someone who would play MS' bill for it, _just to read his mail_. He didn't, originally, do it primarily for the publicity (and note that I'm not saying he didn't _think_ about it...).
I just don't see that a non-free-software-minded person, say an NT applications programmer, would necessarily have had the same reaction -- for that person, it's much more likely to be "just a job"... not a passion.
The HP's are so well supported that everything these days speaks PCL. Unfortunately, HylaFax doesn't. If anyone knows of a free PCL->PS interpreter, they're reading this. Pointers?
>It's a sad testament that, in an effort to protect our children, our society decides to bind their hands and treat them as invalids before they reach a magic age at which point it is assumed they will have the ability to be "responsible".
Indeed. And this is precisely why 'Murrican law on the topic upholds the validity of contracts with minors concerning things essential to living -- as that phrase is interpretd by the judiciary.
IANAL, and I haven't researched it lately, but I think a home phone would count, rent and other utilities certainly do.
I was one of, and possibly the earliest, agitator for Corel to become involved with Linux, back even before the Netwinder project. It's pleasant to see things happening, even if they're not happening as quickly as some people might like to see.
I can't -wait- for Draw for Linux... with that and Ventura, I can dump Windows completely.
> Corel can have the newbie market, and I encourage them to file out the rough edges for them. But don't tell me they all have to be like that. Sometimes complete control is important.
I don't recall that anyone _did_ tell you that all [distributions] had to be [newbiefied]. I'm sure RedHat is not planning to go out of business, just because Corel Linux takes off.
On another front, I suspect you'll find that CL is just as damned configurable as any other install of Debian, if you fire up a shell and know where to go. It will be interesting to see if the configuration tools that Corel has built atop it are as resilient in the face of discovered changes that _they didn't make_, like, say, Linuxconf, but hey...
Whenever a new protocol comes out that purports to solve all your problems, _and there's already protocol out there to do what you need_, examine the motives of the proponents _very_ carefully.
Of course, whether they come out of the chute hot or not, we still owe Transmeta for paying for most of 2.2 and 2.3.;-)
[ comments designed to start fires; yeah; that's the ticket. ]
I expect that there will be stream files of the cast available later. I wonder if someone will hit on the cp command on the webserver precisely at noon...
> This could be interesting ground for the networks. They actually used "Luma" sampling a couple of years ago and got in trouble. I believe it was Dianne Sawyer, Supposedly at the white house or the capital building.
"Chroma Keying". Usually done against a blue, or more commonly these days, green background.
Yes, they did, yes they did get in trouble, an apology was aired on the next newscast. The New York Times covered this in their article on the subject today, in print.
> Anyone else remember Mark Williams Games? They were basically shut down because one of their systems ran a BBS that was used to transfer a description of the 911 system in Georgia.
Yep, that compiler was even more fun than FreeCiv.
:-)
I think you're talking about Steve Jackson Games, maybe? Mark Williams was in... another sector of the computing market.
Immaterial. The Palm VII isn't _using_ TCP. It's a RAM/Mobitex radio. I don't think the overlying layer 2 protocol is IP, I think it's X.25 or something similar.
I'm not positive, you understand, but I do know that Mobitex is _not_ an IP network.
> With disclosed source-code or true Open Source the details are out there where people can see them and it's more likely a problem will be publicly known.
Precisely.
A good case could be made that the coder has taken _extra effort_ in good faith by releasing his source code so that many thousands of eyes could look at it, which would make him or her even less subject to a negligence suit.
Too bad civil defense isn't parallelizable.
Oh, yeah; it is. Who's running that OSS legal defense fund thing?
Slashdot Mirror
Oh, Ghod; I hate this whole "the desktop has no edges" paradigm... (Of course, I hate the word 'paradigm', too, but what the hell... :-)
Am I the only one who thinks this is a _bad_ thing? People want limits, they want walls; this has been proven so many times I can't count them.
Certainly, those walls should not _preclude_ you from finding and making use of the things you need, but really, now; do you _want_ to be unable to discern where that love note to your amor at work... or your HIV test results are; whether that file is on your zip disk, your hard drive, your office fileserver... or just 'out there on the web somewhere'?
Really; c'mon now. Reductio ad absurdum, maybe, but look at it the other way: your web browser, properly configured, is actually your generic file viewer, whether or not those files are _local_, right?
Discipline -- walls -- in a word, are good.
Cheers,
-- jra
-----
These might be good ideas, but I simply don't believe that it's practical to implement them in an environment where "everybody else's" Windows program works some other way.
(This is completely apart from my opinion that these things comprise fundamental computer literacy and that, contrary to Cooper's assertions in Inmates, computer literacy is not a polite way of saying 'stupid willingness to tolerate insufferably poor design -- I don't think HFS' are poor design.)
But, my question would be (yes, there's actually a question buried in here) this: you, Jakob, seem to go the other direction, displaying an almost slavish devotion to the installed base, even if a) it's a poor design, objectively, and b) there's an obviously better one.
Color choices for links would be my favorite example. My weblog, for example, uses a dark teal text color (dark enough for good contrast on white), and navy blue for links -- which drops back to the text color, while still staying underlined, for visited links.
Obviously, this wouldn't work on navigational elements rather than in-line links, but I'm honestly not at all certain that visited-ness is pertinent there, and it might actually be confusing. (I mean, here, the sort of things many people put on 'tabs'.)
I'm an engineering type, so I suppose I'd rather have it 'right', even if it means I have to work harder training people the first time, because if I teach them to fish (discerning rules for link changes from looking at a site) rtaher than feeding them (blue means link, red means visited), then my long-term support load decreases.
Do you think that people, in general, are really so untrainable that this is an unreasonable approach? Or that making them think, even just a little bit, is really that much to ask?
Cheers,
-- jra
-----
although, what's Compaq doing with the Itsy these days... that rock and roll controller is pretty spiffy...
Cheers,
-- jra
-----
Without chasing everyone over to my weblog for the grind, the favorite in the room I was watching in was the Tabasco spot (the 'log has a link to the QT4 file), followed by the Clydesdale spot, the FedEx/Wizard of Oz spot, and the VISA 'swimmers' spot, in differing orders of preference.
We did like the Christopher Reeve one, too. And the "slidewalk/credit card fraud" company's stuff sounded pretty cool, too... who were they again? :-)
Cheers,
-- jra
-----
Piers! Of course!
I, personally, was quite impressed by the 7 books of the "Incarnations of Immortality" series -- primarily because I am _not_ a fantasy reader, by and large. I don't much mind the "Wizard" series from Rick Cook because of all the nifty computing jokes he slips in -- being a long time writer for Byte as well (Toth Set Ra, indeed!).
But the Incarnations, especially books 6 and 7 when he finally got them written, were just hard enough to make me happy.
Cheers,
-- jra
-----
> You know there's actually a law in Florida from the turn of the Century (still in effect) that goes something like:
"When a person driving a motor vehicle reaches a horse and the horse acts frightened, the driver of the vehicle must get out of the car and take it apart, laying its pieces in front of the horse so the horse will calm down"
Um, like hell there is.
I have a set of Florida lawbooks on my office shelf. I've read every word of chapters 316-321, which cover motor vehicles, driving, licensure and titles. Nothing like this exists anywhere in that section.
You have a citation?
(Yeah, I know, he was just trying to make a point... with cobbled-up facts. Furrfu!)
Cheers,
-- jra
-----
The original writer notes:
> But this is not an OS issue, it never was.
I think he's wrong, and our much more than usually eloquent Anonymous Coward appears to agree:
> We can't all work together, because some of "us" depend for "our" existence on keeping code proprietary, and some of us depend on keeping it free. Microsoft can't compromise with free-software, because if it does, it effectively dies as Microsoft. Sure, it could exist as a distro company for FreeWindows2000, but it would no longer be Microsoft in anything but name.
And this comes very close to making the point, for me. It was, roughly, would an NT user/programmer have done this for Microsoft? Or, even moreso, for a Linux company?
This is the same argument, really, as "oh, he's a fashion designer, he must be gay."
No, being gay doesn't make you a better fashion designer. However, _having been gay_ may well predispose you to a greater disdain for society's opinion of whether men should engage in such pursuits. Correlation doesn't imply causation, no. However, that is _not_ sufficient reason to deny such correlations.
The mindset that makes someone likely to hack on Linux, or even use it (still, Corel isn't on 40 million dekstops _yet) is one that's more likely to yield someone who would play MS' bill for it, _just to read his mail_. He didn't, originally, do it primarily for the publicity (and note that I'm not saying he didn't _think_ about it...).
I just don't see that a non-free-software-minded person, say an NT applications programmer, would necessarily have had the same reaction -- for that person, it's much more likely to be "just a job"... not a passion.
Cheers,
-- jra
-----
Since it's only partially related, I've put it here.
The magic word is TANSTAAFL.
Cheers,
-- jra
-----
And this raises a point for me.
The HP's are so well supported that everything these days speaks PCL. Unfortunately, HylaFax doesn't. If anyone knows of a free PCL->PS interpreter, they're reading this. Pointers?
Cheers,
-- jra
-----
I believe this is the logical conclusion of
:-)
snail://USA/33709-4819/63Way/4331
... an address I've always wanted to write on an envelope to see if it got delivered.
(though no doubt the USPS will choose a different protocol identifier...)
Cheers,
-- jra
-----
> I am thinking this will alow me to telnet to my box at my house from school with a cell phone... at least at some point.
Nope.
That's precisely what they'd like for you to think. Read the paper in my above message to see why that's _not_ what will actually happen.
Cheers,
-- jra
-----
>It's a sad testament that, in an effort to protect our children, our society decides to bind their hands and treat them as invalids before they reach a magic age at which point it is assumed they will have the ability to be "responsible".
Indeed. And this is precisely why 'Murrican law on the topic upholds the validity of contracts with minors concerning things essential to living -- as that phrase is interpretd by the judiciary.
IANAL, and I haven't researched it lately, but I think a home phone would count, rent and other utilities certainly do.
Cheers,
-- jra
-----
A nice reply; thanks.
I was one of, and possibly the earliest, agitator for Corel to become involved with Linux, back even before the Netwinder project. It's pleasant to see things happening, even if they're not happening as quickly as some people might like to see.
I can't -wait- for Draw for Linux... with that and Ventura, I can dump Windows completely.
Cheers,
-- jra
Cheers,
-- jra
-----
> Corel can have the newbie market, and I encourage them to file out the rough edges for them. But don't tell me they all have to be like that. Sometimes complete control is important.
I don't recall that anyone _did_ tell you that all [distributions] had to be [newbiefied]. I'm sure RedHat is not planning to go out of business, just because Corel Linux takes off.
On another front, I suspect you'll find that CL is just as damned configurable as any other install of Debian, if you fire up a shell and know where to go. It will be interesting to see if the configuration tools that Corel has built atop it are as resilient in the face of discovered changes that _they didn't make_, like, say, Linuxconf, but hey...
Cheers,
-- jra
-----
Um, "oops".
:-)
And I forgot to make my primary point too. Drat.
Whenever a new protocol comes out that purports to solve all your problems, _and there's already protocol out there to do what you need_, examine the motives of the proponents _very_ carefully.
Cf: DivX.
Cheers,
-- jra
-----
Meant to get this comment in earlier, but I'd lost the link, and Dan Lyke, at Flutterby, whose weblog I'd gotten the link from, hadn't run his index engine recently; I had to mail him for it.
If you have any interest in WAP et. al, at all, I'd suggest reading this paper.
Cheers,
-- jra
-----
Don't Go Here.
Cheers,
-- jra
-----
Of course, whether they come out of the chute hot or not, we still owe Transmeta for paying for most of 2.2 and 2.3. ;-)
[ comments designed to start fires; yeah; that's the ticket. ]
I expect that there will be stream files of the cast available later. I wonder if someone will hit on the cp command on the webserver precisely at noon...
Cheers,
-- jra
-----
"Chroma Keying". Usually done against a blue, or more commonly these days, green background.
Yes, they did, yes they did get in trouble, an apology was aired on the next newscast. The New York Times covered this in their article on the subject today, in print.
Cheers,
-- jra
-----
> Anyone else remember Mark Williams Games? They were basically shut down because one of their systems ran a BBS that was used to transfer a description of the 911 system in Georgia.
Yep, that compiler was even more fun than FreeCiv.
:-)
I think you're talking about Steve Jackson Games, maybe? Mark Williams was in... another sector of the computing market.
Cheers,
-- jra
-----
Actually, it's worse that that; the Germans _asked_ for Hitler to take over.
:-) at baylink.p itas.com.
ESR wrote this up; I've just linked it off my new weblog (there's a shocking idea...
Scary stuff...
Cheers,
-- jra
-----
Actually, it's worse that that; the Germans _asked_ for Hitler to take over.
:-) at baylink.pitas.com.
ESR wrote this up; I've just linked it off my new weblog (there's a shocking idea...
Scary stuff...
Cheers,
-- jra
-----
I was going merely to pooh-pooh this comment... until I saw who'd written it. You're perhaps the only person I'll accept that opinion from.
:-)
But where do you come by it?
And where has your website gone?
Cheers,
-- jra
-----
> Anybody know how the Palm VII's TCP/IP is?
Immaterial. The Palm VII isn't _using_ TCP. It's a RAM/Mobitex radio. I don't think the overlying layer 2 protocol is IP, I think it's X.25 or something similar.
I'm not positive, you understand, but I do know that Mobitex is _not_ an IP network.
Cheers,
> With disclosed source-code or true Open Source the details are out there where people can see them and it's more likely a problem will be publicly known.
Precisely.
A good case could be made that the coder has taken _extra effort_ in good faith by releasing his source code so that many thousands of eyes could look at it, which would make him or her even less subject to a negligence suit.
Too bad civil defense isn't parallelizable.
Oh, yeah; it is. Who's running that OSS legal defense fund thing?
Cheers,